INDEX

Explanations

feature extraction and selection

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hitva

1.92

enação

1.81

oyeva

1.80

etor

1.76

utiérrez

1.75

ेट्टी

1.71

iado

1.70

atthena

1.70

ojana

1.70

akkhati

1.70

POSITIVE LOGITS

 Rounds

1.95

Pron

1.54

Rounded

1.53

 Pron

1.50

 Schedule

1.49

Spacing

1.47

 Rounded

1.47

 Speaks

1.45

$^{

1.42

Plate

1.42

Activations Density 0.003%